Re: [vpp-dev] How to actively close the client connections in http server? #vnet

. So, apart from calling vnet_disconnect_session () on an rx event, have you done anything else to the code? Regards, Florin > On Nov 21, 2018, at 1:36 AM, Andreas Schultz > wrote: > > Florin Coras mailto:fcoras.li...@gmail.com>> schrieb > am Fr., 18. Mai 2018 um 08:12 Uh

Re: [vpp-dev] VPP and TCP half-open connection

Hi Rubina, That’s a tcp protocol startup configuration parameter that can be used to preallocate half-open connections and therefore prevent the underlying pool from expanding. It’s especially useful if you expect applications attached to vpp’s host stack to open a large number of tcp

Re: [vpp-dev] VPP and TCP half-open connection

If your question concerns the host stack, the answer is no. Florin > On Nov 19, 2018, at 5:08 AM, Rubina Bianchi wrote: > > Hi Dear VPP > > Is there any mechanism in VPP to restrict TCP half-open connection count? If > answer is yes how it can be configured? > > Thanks, > Sincerely >

Re: [vpp-dev] test-ext failures seen on master

a binary called sock_test_client is > left running after a vpp crash or test failure. What's worse, it eats > 100% CPU. > > Quoting Florin Coras (2018-11-16 00:40:06) >> Thanks, Dave! >> I’ll take a look at those as soon as I can. I’m running multiple >> connec

Re: [vpp-dev] test-ext failures seen on master

That’s an interesting failure. Is the test machine running out of memory? The extended tests are unstable on my server, so I do see quite a number of failures. However this: make test-ext TEST=vcl.VCLCutThruTestCase.test_vcl_cut_thru_uni_dir_nsock runs just fine. After the latest test

Re: [vpp-dev] inet raw socket

Option 3.5, if one plans to use any of the transport protocols, would be to write a builtin application, as opposed to using vcl/raw session layer apis. Florin > On Nov 12, 2018, at 11:47 PM, Ole Troan wrote: > > Hi again, > >> yes, we have a control plane, and the protocols I mentioned

Re: [vpp-dev] #vpp sctp echo test crash

s)?? > Thoughts? > > Marco Varlese > > On Nov 8, 2018 22:06, Florin Coras wrote: > Hi Marco, > > So echo test with 500MB does finish and reports throughput after 15s but with > warnings of the type “clients stuck”? > > Note that we also have tests that run iper

Re: [vpp-dev] #vpp sctp echo test crash

hat happens for SCTP. I did try to increase the test-timeout > but without much success... > > Interestingly, for both TCP and SCTP, with a test-timeout of 30 seconds the > test completes (with errors) after round 12-15 secs... > > On Thu, 2018-11-08 at 08:21 -0800, Florin Cor

Re: [vpp-dev] #vpp sctp echo test crash

Hi guys, I unfortunately missed this. Yes, the fifo size provided is the number of kB and yes you should be able to do 4MB (fifo-size 4096) or even higher. The “stuck clients” warning is just a notification, the test can still finish. To make sure you give the test enough time, add

Re: [vpp-dev] VPP and StrongSwan: Issues seen with both master and 18.10 such that StrongSwan crashes when calling vl_client_api_map()

Hi Kyle, Could you try doing something like "clib_mem_init_thread_safe (0, 256 << 20);” kernel_vpp_plugin_create() before calling vac_create()? Hope this helps, Florin > On Nov 5, 2018, at 10:22 AM, Kyle Mestery (kmestery) via Lists.Fd.Io > wrote: > > Greetings! > > I’ve got a fairly

Re: [vpp-dev] inet raw socket

We don’t support raw sockets. Florin > On Nov 5, 2018, at 1:15 AM, saint_sun 孙 via Lists.Fd.Io > wrote: > > > <81.gif>thanks very much! And I have another question: does VPP support INET > RAW socket, I find the UDP and TCP socket except the raw socket. > > saint_...@aliyun.com

Re: [E] [vpp-dev] VPPSB Netlink build failing

Agreed with Damjan. Why not manage this as a separate project, just like Sweetcomb? Goal for Sandbox was to host plugins that could later convert into fully fledged projects. Also, can’t netlink/router plugins use the binary APIs instead of, I would assume, raw C FIB APIs? That should

Re: [vpp-dev] VPP and RCU?

> On Oct 31, 2018, at 4:03 PM, Stephen Hemminger > wrote: > > On Wed, 31 Oct 2018 00:24:36 -0700 > Florin Coras wrote: > >>>>>> No reader-writer locks are 100's of times slower. In fact reader >>>>>> write locks are slower than no

Re: [vpp-dev] VPP and RCU?

> On Oct 30, 2018, at 9:02 PM, Honnappa Nagarahalli > wrote: > > > > Hi Stephen, > > No, we don’t support RCU. Wouldn’t rw-locks be enough to support your >> usecases? > > Florin > >> On Oct 29, 2018, at 12:40 PM, Stephen Hemminger >> wrote: >>

Re: [vpp-dev] VPP and RCU?

> On Oct 29, 2018, at 1:42 PM, Stephen Hemminger > wrote: > > On Mon, 29 Oct 2018 13:20:27 -0700 > Florin Coras wrote: > >> Hi Stephen, >> >> No, we don’t support RCU. Wouldn’t rw-locks be enough to support your >> usecases? >> >>

Re: [vpp-dev] VPP and RCU?

Hi Stephen, No, we don’t support RCU. Wouldn’t rw-locks be enough to support your usecases? Florin > On Oct 29, 2018, at 12:40 PM, Stephen Hemminger > wrote: > > Is it possible to do Read Copy Update with VPP? Either using Userspace RCU > (https://librcu.org) > or manually. RCU is very

Re: [vpp-dev] [csit-dev] VPP 18.10 is out!!!

Awesome! Thanks, Marco! Florin > On Oct 24, 2018, at 12:46 AM, Marco Varlese wrote: > > Dear all, > > I am very happy to announce that release 18.10 is available. > > Release artificats can be found on both Nexus and Packagecloud. > > Thanks to all contributors for yet another great

Re: [vpp-dev] VPP crash after days of heavy traffic, next_index is not associated to an interface #vpp_stability

Perfect! Thanks for confirming! Florin > On Oct 21, 2018, at 12:16 PM, JB wrote: > > Hello Florin, > > Brilliant, that does seem to address it. > > We'll patch it tomorrow. > > Much appreciated. > > Thanks, > John -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this

Re: [vpp-dev] VPP crash after days of heavy traffic, next_index is not associated to an interface #vpp_stability

The issue should be fixed by this [1]. Thanks, Florin [1] https://gerrit.fd.io/r/#/c/15430/ > On Oct 21, 2018, at 7:23 AM, Dave Barach via Lists.Fd.Io > wrote: > > Thanks for the report… This is the sort of thing we need to clean up. We do > very little negative testing of the form: send

Re: [vpp-dev] Elog binary file format

As for the crash, try right at the start of your main: clib_mem_init_thread_safe (0, 256 << 20) Florin > On Oct 9, 2018, at 4:55 AM, Aleksander Djuric > wrote: > > Correction: I've found all of unserialize_* functions. Trying to understand.. > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive

Re: [vpp-dev] Elog binary file format

Also, take a look at c2cpel and cpeldump. The first allows you to convert log files to cpel files while the second converts cpel to text. Hope this helps, Florin > On Oct 8, 2018, at 9:22 AM, Damjan Marion via Lists.Fd.Io > wrote: > > > -- > Damjan > >> On 8 Oct 2018, at 07:37,

Re: [vpp-dev] Error when run 'socket_test.sh -bi docker-preload' for vcl

Hi Yalei, I'm exclusively test using vcl_client/server and sock_test_client/server because it’s much easier to debug. At first glance, it looks as if vpp_eth_name is not properly configured in the script. Maybe Dave can provide more details. Florin > On Sep 20, 2018, at 5:18 AM, wylandrea

Re: [vpp-dev] build queue backed up a bit foo

jobs have finished more than one and half-hour ago and the patch >>> cannot be >>> marked Verified+1 because Jenkins is still waiting for the ARM job to >>> complete >>> (timeout = 120 minutes). IMHO it makes the overall patch submission and >>> review >

Re: [vpp-dev] build queue backed up a bit foo

ARM jobs have not been working for some days now. That’s why their result is skipped. Timeout is 2h but probably we should drop that even further … Florin > On Sep 7, 2018, at 11:59 AM, Ole Troan wrote: > > Trying out a change in: > https://gerrit.fd.io/r/#/c/14732/

Re: [vpp-dev] TLS half open lock

we still need to use lock to > resolve this race condition. > >   <> > <>From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Yu, > Ping > Sent: Tuesday, August 28, 2018 5:

Re: [vpp-dev] TLS half open lock

; thread 0: alloc: 146 > thread 2 free: 153 > thread 0: alloc: 144 > thread 0: alloc: 153 > thread 0: alloc: 124 > thread 3 free: 25 > thread 0: alloc: 25 > > From: Florin Coras [mailto:fcoras.li...@gmail.com] > Sent: Tuesday, August 28, 2018 10:24 AM > To: Yu, Pin

Re: [vpp-dev] TLS half open lock

vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Florin > Coras > Sent: Tuesday, August 28, 2018 12:51 AM > To: Yu, Ping > Cc: Florin Coras (fcoras) ; vpp-dev@lists.fd.io > Subject: Re: [vpp-dev] TLS half open lock > > Hi Ping, > > The current imp

Re: [vpp-dev] TLS half open lock

Hi Ping, The current implementation only locks the half-open pool if the pool is about to expand. This is done to increase speed by avoiding unnecessary locking, i.e., if pool is not about to expand, it should be safe to get a new element from it without affecting readers. Now the thing to

Re: [vpp-dev] Control of transit TCP sessions

Hi Aleksander, The session layer session lookup tables lower are used for transport connections that are terminated in vpp. They can be used to store any connection but, as you’ve already figured out, they’ve been written to work with sessions and transport connections. Moreover, session

Re: [vpp-dev] TCP retransmission

gt;   <> > <>From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Florin > Coras > Sent: Friday, August 24, 2018 9:37 AM > To: Yu, Ping > Cc: vpp-dev@lists.fd.io; Florin Coras (fcoras) > Subject: Re: [vpp-dev] TCP retransmission > > Ping, >

Re: [vpp-dev] TCP retransmission

2 > > thanks > Ping > > > From: Yu, Ping > Sent: Thursday, August 23, 2018 4:49 PM > To: Florin Coras > Cc: vpp-dev@lists.fd.io; Florin Coras (fcoras) ; Yu, Ping > > Subject: RE: [vpp-dev] TCP retransmission > > Verbose 1 also shows some info belo

Re: [vpp-dev] Crash when using the debug vpp image

Hi Sirshak, Try: sudo rm -rf /dev/shm/* Florin > On Aug 23, 2018, at 9:09 AM, Sirshak Das wrote: > > Hi all, > > I am trying to debug a segfault due to of my commits. I usually use the > release image by building make build-release, but as I am trying to debug, so > I decided to just do

Re: [vpp-dev] TCP retransmission

Hi Ping, Just from this, I can’t tell precisely what the issue is. Once the test hangs, do “show session verbose 2” and you should get the detailed status of the sessions that froze. I’ll take a look at those once I’m back in the office. Retransmissions should work fine but one never knows.

Re: [vpp-dev] Where could I find the example of vcl.conf

> On Aug 16, 2018, at 4:12 AM, 汪亚雷 wrote: > > Thanks Florin, > > It works fine now. :) > And two more questions: > 1. Will eventfd replace the condvar way? it will be the default way for event > queue? > 2. Will socket api replace the vpe api in the future? > > Than

Re: [vpp-dev] Where could I find the example of vcl.conf

seg } > #socksvr { socket-name /run/vpp-api.sock } > > == > root@192.168.122.252:/home/wenjiang/dmm/release/bin (master) $ cat > /etc/vpp/vcl.conf > vcl { > #rx-fifo-size 400 > #tx-fifo-size 400 > app-scope-local > app-scope-global

Re: [vpp-dev] Where could I find the example of vcl.conf

Hi Yalei, You have an example of how to write a vcl.conf file in vcl/vcl_test.conf. It’s just an example, so if you want to try out eventfd, here’s what I’ve been recently using: vcl { rx-fifo-size 400 tx-fifo-size 400 app-scope-local app-scope-global api-socket-name

Re: [vpp-dev] tls init server is too heavy

:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Yu, > Ping > Sent: Thursday, July 26, 2018 9:26 AM > To: Florin Coras (fcoras) mailto:fco...@cisco.com>>; > vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > Cc: Yu, Ping mailto:ping...@intel.com>&

Re: [vpp-dev] vnet tcp.h include error

Hi Korian, Does this [1] solve your problem? Florin [1] https://gerrit.fd.io/r/#/c/14178/ > On Aug 13, 2018, at 7:36 AM, ko wrote: > > Hello Dave, > > I get the same error with 18.07 (from >

Re: [vpp-dev] About the fifo-size parameter in app tcp_echo

Hi Yalei, I guess there’s two solutions to that. Set min_chunk to something like clib_min (16 << 10, fifo_size/2) or remove min_chunk entirely. I’m fine with either of those. Feel free to push a patch! Thanks, Florin > On Aug 8, 2018, at 12:17 AM, 汪亚雷 wrote: > > Hi Florin, > > These days

Re: [vpp-dev] Assertion fails in external echo server/client

ersion last time. > > > > - > > YeXin > > > > 发件人: Florin Coras [mailto:fcoras.li...@gmail.com > <mailto:fcoras.li...@gmail.com>] > 发送时间: 2018年8月6日 23:35 > 收件人: yexin (F) mailto:yexi...@huawei.com>> > 抄送: vpp-dev@lists.f

Re: [vpp-dev] Assertion fails in external echo server/client

Hi YeXin, First of all, do you see this with 18.07 or master? Secondly, I’ve checked out origin/stable/1804 and at line 1126 there’s no attempt to access a pool. Are you testing with rc0 or custom code? Thanks, Florin > On Aug 5, 2018, at 11:29 PM, yexin (F) wrote: > > The version tested

Re: [vpp-dev] About the vppcom_epoll_wait function in vcl

Hi Yalei, Are you testing with master latest? Yes, blocking mode will be properly supported. The reason it won’t block now is because we need to poll multiple message queues (vpp’s and cut-through sessions) and -1 is not interpreted as a large number. This will be fixed once we switch from

Re: [vpp-dev] Regarding VPP TCP Stack usage

ng penalty due to > the need to handle both kernel socket based file descriptors and VCL/VPP > created file descriptors with the LD_PRELOAD callback functions. > > Thanks, > -daw- > > On 7/31/18 2:11 PM, Florin Coras wrote: >> Hi Matt, >> >> I’d say that

Re: [vpp-dev] Control TCP_NODELAY and TCP_QUICKACK per session

Hi Tomasz, We actually have both hardwired to “on" in the the stack. The reason why they’re on VCL’s list is because we don’t have APIs to inspect/modify them. There’s no plan to support Nagle’s algorithm but we should support at some point adaptive acks. Also, in case you haven’t been

Re: [vpp-dev] [discuss] vpp region init fail Couldn't connect to vpe, exiting...

Probably the right list would be vpp-...@fd.io . As for your question, is vpp actually up and running? What does ps -ef | grep vpp show? Florin > On Aug 1, 2018, at 10:40 AM, jkzcristiano wrote: > > Hi all, > > This is the first time I write to this group and I am

Re: [vpp-dev] Regarding VPP TCP Stack usage

hanks! > -Matt > > >> On Jul 30, 2018, at 10:39 AM, Florin Coras > <mailto:fcoras.li...@gmail.com>> wrote: >> >> Prashant, >> >> Dave is exactly right. If you still want to try out the LDP layer, I >> wouldn’t set a global LD_PRELOA

Re: [vpp-dev] Regarding VPP TCP Stack usage

Prashant, Dave is exactly right. If you still want to try out the LDP layer, I wouldn’t set a global LD_PRELOAD variable because that will end up preloading all the applications and, inevitably, to some unsupported usage patterns and crashes. Instead, start only your app with LD_PRELOAD set,

Re: [vpp-dev] vcl error

t;is_ip4, ep->ip); > > // to_ip46 first parameter is isipv6, and IP46_TYPE_IP4, IP6 would be always > true, and the bind will copy a wrong IP for v4 case(over stack) > } > > BR//Terry > > -- Original -- > From: "Florin Coras";

Re: [vpp-dev] vcl error

Hi Terry, From the errors you’re seeing, I can only guess you have configured vcl to attach to the host stack using a local scope. Do you have vcl.conf configured with “app-scope-local” or have you exported VCL_APP_SCOPE_LOCAL environment variable? If yes, try disabling them. If neither

Re: [vpp-dev] UDP echo server does not work properly with IPsec enabled

Hi Berenger, Thanks for the report! There are two issues here. First of all, echo_server/client apps are testing apps that can misbehave is not properly configured. I just tried master with 4 workers for both and with the following configuration: server: test echo server fifo-size 4096 uri

Re: [vpp-dev] ONE topology not working #vnet #vpp

Hi Saurabh, Glad to see you’re interested in LISP! As the error lower suggests, you first need to enable the lisp code. You can achieve this by doing: “one enable". Note that although “lisp enable” would also have the same result, going forward we plan to deprecate the “lisp” apis in favor

Re: [vpp-dev] FD.io VPP Docs (Devices)

Out of pure curiosity - was SSVM supported by VPP in the past and the > support was removed ? Or > was it never supported by VPP ? > > Regards, > Rami Rosen > > > On 6 July 2018 at 19:58, Florin Coras wrote: >> We don’t have ssvm devices anymore. >> >> F

Re: [vpp-dev] FD.io VPP Docs (Devices)

We don’t have ssvm devices anymore. Florin > On Jul 6, 2018, at 9:21 AM, Ray Kinsella wrote: > > Hi Folks, > > The FD.io Docs WG and I are working through the list of FD.io VPP features. > We are hoping to crowd source help by asking specific questions (and > hopefully getting some

Re: [vpp-dev] gerrit.fd.io very slow?

It’s slow on my end as well. Florin > On Jun 20, 2018, at 8:21 AM, Dave Barach via Lists.Fd.Io > wrote: > > Before I start bugging the LF: is anyone else noticing bad performance w/ > gerrit.fd.io ? > > Thanks... Dave > > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive

Re: [vpp-dev] VLIB_BUFFER_TOTAL_LENGTH_VALID set but invalid total_length_not_including_first_buffer

Hi Yoann, Shouldn’t the clone code only read s->total_length_not_including_first_buffer only if VLIB_BUFFER_NEXT_PRESENT flag is set for s? I’m thinking about this line in particular: d->total_length_not_including_first_buffer = s->total_length_not_including_first_buffer +

Re: [vpp-dev] [FD.io Helpdesk #56282] git.fd.io not updating

>> Roughly a week ago, I noticed there was a DNS/IP change when cloning a >> new VPP >> repo... I wonder if what I saw is somehow connected to this issue. >> On Mon, 2018-05-21 at 16:34 -0700, Florin Coras wrote: >>> Hi, >>> It would seem that git.fd.io [1]

Re: [vpp-dev] [FD.io Helpdesk #56282] git.fd.io not updating

>> Roughly a week ago, I noticed there was a DNS/IP change when cloning a >> new VPP >> repo... I wonder if what I saw is somehow connected to this issue. >> On Mon, 2018-05-21 at 16:34 -0700, Florin Coras wrote: >>> Hi, >>> It would seem that git.fd.io [1]

Re: [vpp-dev] git.fd.io not updating

: >>> Roughly a week ago, I noticed there was a DNS/IP change when cloning a new >>> VPP repo... I wonder if what I saw is somehow connected to this issue. >>> >>> On Mon, 2018-05-21 at 16:34 -0700, Florin Coras wrote: >>>> Hi, >>>> >

Re: [vpp-dev] Is there a plan to support the udp mode in vcl?

Xyxue, It’s on our todo list. In the mean time, you can use the raw session layer apis. For examples see src/tests/vnet/session/udp_echo.c Florin > On May 21, 2018, at 11:08 PM, xyxue wrote: > > Hi Florin， Ed， > > I'm testing the vcl in udp mode. After our debug vpp

[vpp-dev] git.fd.io not updating

Hi, It would seem that git.fd.io [1] thinks that we last committed a patch to vpp almost 1 week ago. Any idea what might’ve triggered this? Thanks, Florin [1] https://git.fd.io/vpp/log/

Re: [vpp-dev] How to actively close the client connections in http server? #vnet

That http server is just example code that executes the contents of a get request as a cli commands within a spawned vpp process. So, if you want to disconnect _after_ the the reply is sent, call vnet_disconnect_session () at the end of http_cli_process. Florin > On May 17, 2018, at 10:52 PM,

Re: [vpp-dev] gerrit.fd.io froblem access

IN CNAME awscloud.fd.io. awscloud.fd.io. 5 IN A 52.10.107.188 Florin > On May 15, 2018, at 3:03 PM, Florin Coras <fcoras.li...@gmail.com> wrote: > > I’m getting in Chrome NET::ERR_CERT_COMMON_NAME_INVALID > > And dig says: &

Re: [vpp-dev] gerrit.fd.io froblem access

I’m getting in Chrome NET::ERR_CERT_COMMON_NAME_INVALID And dig says: ;; ANSWER SECTION: gerrit.fd.io. 5 IN CNAME dev.fd.io. dev.fd.io. 5 IN A 162.253.54.31 Florin > On May 15, 2018, at 2:50 PM, Dave Barach wrote: > >

Re: [vpp-dev] question about the VCL

Xyxue, VCL hasn’t yet been updated to work in dgram mode. Try the builtin udp echo client/server or tests/vnet/session/udp_echo. Note however that you can’t expect udp to deliver all the bytes from one end to the other because of packet drops so try half-duplex testing. For some examples see

Re: [vpp-dev] Using custom openssl with vpp #vpp

Hi DucTM, Did you try changing src/plugin/tlsopenssl.am to link against openssl 1.1? I’ve never tried it, so no idea what the end result may be :-) Florin > On May 14, 2018, at 3:52 AM, duct...@viettel.com.vn wrote: > > Hi, > I'm trying to customize the openssl plugin that needs to work with

Re: [vpp-dev] question about the VCL

Hi Xyxue, No, the stack does not support IPPROTO_RAW. Given that this is a user space stack and that you have access to things like memif, may I ask what use case you would need that for? Florin > On May 14, 2018, at 12:58 AM, xyxue wrote: > > > Hi guys, > > Is the

Re: [vpp-dev] TCP performance - TSO - HW offloading in general.

ust be the reason. > > Luca > > > > From: Florin Coras <fcoras.li...@gmail.com> > Date: Friday 11 May 2018 at 23:02 > To: Luca Muscariello <lumuscar+f...@cisco.com> > Cc: "vpp-dev@lists.fd.io" <vpp-dev@lists.fd.io> > Subject: Re: [vpp-dev] T

Re: [vpp-dev] TCP performance - TSO - HW offloading in general.

ay-based congestion control does not like > much VPP batching (batching in general) and we are using DBCG. > > Linux TSO has the same problem but has TCP pacing to limit bad effects of > bursts > on RTT/losses and flow control laws. > > I guess you’re aware of these issues al

Re: [vpp-dev] TCP performance - TSO - HW offloading in general.

Hi Luca, We don’t yet support TSO because it requires support within all of vpp (think tunnels). Still, it’s on our list. As for crypto offload, we do have support for IPSec offload with QAT cards and we’re now working with Ping and Ray from Intel on accelerating the TLS OpenSSL engine also

Re: [vpp-dev] segfault due to movaps unaligned access

@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Florin > Coras > Sent: Thursday, April 26, 2018 4:13 PM > To: Nicolau, Radu <radu.nico...@intel.com <mailto:radu.nico...@intel.com>> > Cc: vpp-dev@lists.fd.io <mailt

Re: [vpp-dev] segfault due to movaps unaligned access

ing? > > Regards, > Radu >   <> > <>From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Florin > Coras > Sent: Tuesday, April 24, 2018 11:25 PM > To: Nicolau,

Re: [vpp-dev] [csit-dev] VPP 18.04 has been released! Artifacts are available on nexus.fd.io

Yay! Thanks, Chris! Florin > On Apr 25, 2018, at 3:52 PM, Chris Luke wrote: > > All, > > I am happy to announce that VPP 18.04 has been released! The release notes > can be found athttps://docs.fd.io/vpp/18.04/release_notes_1804.html >

Re: [vpp-dev] segfault due to movaps unaligned access

Hi Radu, Making the crypto_worker_main_t a full cache line in size (see patch [1]) seems to solve the issue. Could you confirm? Florin [1] https://gerrit.fd.io/r/#/c/12086/ > On Apr 24, 2018, at 9:23 AM, Radu Nicolau wrote: > >

Re: [vpp-dev] mbedtls in 18.04

Hi Tom, Feel free to remove it as a hard dependency, since the infra should handle it. We now also have support for an openssl engine and as far as I remember, libssl is a hard dependency for the rpm builds. Florin > On Apr 5, 2018, at 5:05 AM, Thomas F Herbert wrote:

Re: [vpp-dev] 18.04 RC1 is cut, master is open again

Awesome! Thanks, Chris! Florin > On Apr 4, 2018, at 7:40 PM, Chris Luke wrote: > > All, > > The stable/1804 branch was created earlier; master is now open again for > general patches. > > I’m not ready to accept fixes on the new stable branch. This is pending >

Re: [vpp-dev] VCL UDP connection failure

pcom.c:3185 > #6 0x77a9f1e3 in vppcom_session_recvfrom > (session_index=session_index@entry=0, buffer=buffer@entry=0x7fffdf40, > buflen=buflen@entry=1024, flags=flags@entry=0, > ep=ep@entry=0x7fffde10) at /vpp/build-data/../src/vcl/vppcom.c:5602 > #7 0x7

Re: [vpp-dev] VCL UDP connection failure

ttp://0.0.0.0:0/>ldp-2519-app > > > Thanks, > Satish. > > > On Thu, Mar 29, 2018 at 5:49 PM, Florin Coras <fcoras.li...@gmail.com > <mailto:fcoras.li...@gmail.com>> wrote: > Hi Satish, > > What does “show app server” report? > > Than

Re: [vpp-dev] VCL UDP connection failure

Hi Satish, What does “show app server” report? Thanks, Florin > On Mar 29, 2018, at 3:14 PM, Satish wrote: > > Hi, > > I am trying to use Host stack of VPP for UDP connection through VCL library. > Server application listen on port 7891. > > But when i try to

Re: [vpp-dev] session handlers and UDP packet boundaries?

layer and the existing apps the expect datagram service. If you’re still interested in more details, let’s take this off list. Florin > On Mar 28, 2018, at 4:36 PM, Andreas Schultz <andreas.schu...@travelping.com> > wrote: > > Hi Florin, > > Florin Coras

Re: [vpp-dev] session handlers and UDP packet boundaries?

Hi Andreas, Not at this time. It’s on our todo list. Regards, Florin > On Mar 26, 2018, at 12:11 PM, Andreas Schultz > wrote: > > Hi, > > The session app handler converts everything into a byte stream, even datagram > protocols like UDP. Is there a way to

Re: [vpp-dev] UDP ECHO Server

> > [root@vbbubng6-shashi-7 vpp]# ./build-root/build-vpp-native/vpp/udp_echo > slave uri udp://192.168.180.30/5678 use-svm-api > (nil): usage [master|slave] > [root@vbbubng6-shashi-7 vpp]# > > > What could be the issue with UDP? > > Regards > Shashi >

Re: [vpp-dev] UDP ECHO Server

for msg id 21 > msg_handler_internal:432: no handler for msg id 21 > > > regards > Shashi > > > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Florin > Coras > Sent: 23 March 201

Re: [vpp-dev] UDP ECHO Server

t; > regards > Shashi > > > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Florin > Coras > Sent: 23 March 2018 14:42 > To: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> >

Re: [vpp-dev] UDP ECHO Server

0 snd_congestion 2735098422 dupack 0 limited_transmit 2735098422 > tsecr 0 tsecr_last_ack 0 > rto 6 rto_boff 9 srtt 0 rttvar 0 rtt_ts 0 rtt_seq 1559868875 > tsval_recent 0 tsval_recent_age 87918962 > > > regards > Shashi > > > > From: vpp-dev@lists.fd.io <mailto:vpp-

Re: [vpp-dev] UDP ECHO Server

Try master. Florin > On Mar 23, 2018, at 8:23 AM, Shashi Kant Singh <sksi...@altiostar.com> wrote: > > What would be the branch I can try this? > > Regards > Shashi > > > From: Florin Coras [mailto:fcoras.li...@gmail.com > <mailto:fcoras.li...@g

Re: [vpp-dev] UDP ECHO Server

alidate http ip6 kube-proxyl2fib > l2patch lb > lookupmfib one policer proxy > session tcp > vpp# test > > > > I cannot find option to run “echo”

Re: [vpp-dev] UDP ECHO Server

I understand: > > vpp# test udp server uri udp://192.168.180.31/5678 > > > Is this correct? > > If so, then, it does not give “udp” when I search under test cli > > > Regards > Shashi > > From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf

Re: [vpp-dev] UDP ECHO Server

Your uri asks for tcp not udp transport service. Florin > On Mar 22, 2018, at 8:25 AM, Shashi Kant Singh wrote: > > Hi > > In the VPP Rel 18:01, I could run the TCP ECHO server as > vpp# test tcp server uri tcp://192.168.180.31/5678 > > But I could not find the udp

Re: [vpp-dev] Error while running TCP_ECHO in standalone mode

The error says it all. You don’t have the session layer enabled. Do “session enable” from VPP’s cli. Florin > On Mar 22, 2018, at 8:57 AM, Shashi Kant Singh wrote: > > Hi > > I could verify TCP_ECHO as the VPP CLI app but when I run as standalone linux > app then I

Re: [vpp-dev] vcl_test_server/client for UDP

Because after enqueueing, the buffers are dropped with error “successfully enqueued”. Check “sh error”. Florin > On Mar 17, 2018, at 1:03 AM, Shashi Kant Singh wrote: > > Hi > > I changed the vcl_test_server / client from TCP to UDP. > > Question: Why does the UDP

Re: [vpp-dev] Error during startup

sts.fd.io> > Subject: Re: [vpp-dev] Error during startup > > +1. Please be careful. This is a cockpit error. > > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> <vpp-dev@lists.fd.io > <mailto:vpp-dev@lists.fd.io>> On Behalf Of Florin Coras > Sent: Thursd

[vpp-dev] Draft vpp host stack wiki page

This took much longer than expected but .. better late than never :-) https://wiki.fd.io/view/VPP/HostStack This is work in progress so expect updates. Regards, Florin

Re: [vpp-dev] VPP As A Router Between Namespaces - 10ms latency

Great! :-) Florin > On Mar 8, 2018, at 1:42 AM, Sara Gittlin wrote: > > Florin > Thank you - i'm done - i get very good numbers with tap both for > throughput and latency > > -Sara > > On Thu, Mar 8, 2018 at 10:55 AM, Sara Gittlin

Re: [vpp-dev] Error during startup

Hi Shashi, This can’t possibly be part of 17.10 and 18.01 since it was only merged this week on master. Probably you need to clean your repo, rebuild and reinstall debs. Florin > On Mar 7, 2018, at 7:44 PM, Shashi Kant Singh wrote: > > Hi > > When I try to start vpp

Re: [vpp-dev] TCP Proxy Through-put

Hi Shaun, Glad to see you’re experimenting with the proxy code. Note however that the implementation is literally just a proof of concept. We haven’t spent any time optimizing it. Nonetheless, it would be interesting to understand why this happens. Does the difference between apache and vpp

Re: [vpp-dev] VPP As A Router Between Namespaces - 10ms latency

Could you try again with taps and large rings? create tap rx-ring-size 4096 tx-ring-size 4096 create tap rx-ring-size 4096 tx-ring-size 4096 Configure then the two taps in your namespaces and run iperf again. Hope this helps, Florin > On Mar 7, 2018, at 11:01 AM, Sara Gittlin

Re: [vpp-dev] number of csit-vpp jobs failing on vpp master branch - RCA wip

Can it be that you're building the packages with tls support and afterwards you’re installing them on a host without libmbedtls-dev? Florin > On Mar 6, 2018, at 7:51 AM, Marek Gradzki -X (mgradzki - PANTHEON > TECHNOLOGIES@Cisco) wrote: > > + hc2vpp (all hc2vpp csit jobs

Re: [vpp-dev] API and use of VLAs

In case of memclnt that’s actually a void pointer to be passed to the rpc handler. Unless it complicates life for the API generator, not sure if it’s worth adding the size to the struct. Florin > On Mar 1, 2018, at 9:26 AM, Ole Troan wrote: > > Guys, > > I'd like to

Re: [vpp-dev] Developing application on top of VPP

Shashi, You can’t really compare macswap with a full fledged host stack. As you delegate more work to vpp, you should expect a drop in throughput. Also worth noting is that the stack sends/receives byte streams, not single packets and up to transport layer, it does make use of vectorization.

Re: [vpp-dev] Errors on compiling NSH plugin with VPP on ubuntu 16.04

Hi Deepak, This should be directed at the nsh dev list. Florin > On Feb 27, 2018, at 9:32 AM, .ılı.D'p@k.ılı. wrote: > > Hi, > > I am getting compilation errors with NSH plugin. > I am using the latest VPP and NSH plugin > And also i downgraded and tried with VPP

<    1   2   3   4   5   6   7   8   >